Pandas DataFrame合并问题：如何基于KO-ST字段合并df1与df2

云原生阿岚

2026-5-25

Pandas Merge Solution for df1 and df2

Hey there! Let's work through merging your two DataFrames (df1 and df2) using Pandas. First, let's recap their structures based on your examples to make sure we're on the same page:

Sample DataFrame Structures

import pandas as pd

# df1 sample
df1 = pd.DataFrame({
    'KO-ST': ['1976-_', '991-_'],
    '1_UID': [200106897, 200108737],
    '2_Vloge': [200106897.0, 200108737.0]
})

# df2 sample
df2 = pd.DataFrame({
    'a': [10000002, 10000003],
    'b': ['851-601', '851-1'],
    'SID': [288.0, 68.0],
    'KO-ST': [288.0, 68.0]
})

A quick note: I noticed KO-ST is stored as strings in df1 but numeric floats in df2. This will break matching—you need to standardize the data type first! For example:

# Convert df1's KO-ST to numeric (handle non-numeric values with 'coerce')
df1['KO-ST'] = pd.to_numeric(df1['KO-ST'], errors='coerce')

# OR convert df2's KO-ST to string
df2['KO-ST'] = df2['KO-ST'].astype(str)

Core Merge Operations

Pandas' pd.merge() is the go-to tool here. We'll use the shared KO-ST column as our merge key, and adjust the how parameter to get the type of join you need:

1. Inner Join (Default)

Keep only rows where KO-ST exists in both DataFrames:

merged_inner = pd.merge(df1, df2, on='KO-ST')

2. Left Join

Keep all rows from df1, and match with df2 rows where possible (missing df2 values get filled with NaN):

merged_left = pd.merge(df1, df2, on='KO-ST', how='left')

3. Right Join

Keep all rows from df2, and match with df1 rows where possible:

merged_right = pd.merge(df1, df2, on='KO-ST', how='right')

4. Outer Join

Keep every row from both DataFrames, filling gaps with NaN where there's no match:

merged_outer = pd.merge(df1, df2, on='KO-ST', how='outer')

Extra Tips for Clean Merges

Handle duplicate column names: If other columns share names across df1/df2, add suffixes to tell them apart:
```
merged_with_suffixes = pd.merge(df1, df2, on='KO-ST', suffixes=('_from_df1', '_from_df2'))
```

Clean messy keys: If KO-ST has hidden whitespace, strip it first:

df1['KO-ST'] = df1['KO-ST'].astype(str).str.strip()
df2['KO-ST'] = df2['KO-ST'].astype(str).str.strip()

Debug missing matches: Find rows in df1 that don't have a corresponding KO-ST in df2:
```
unmatched_df1 = df1[~df1['KO-ST'].isin(df2['KO-ST'])]
```

内容的提问来源于stack exchange，提问作者energyMax